*For correspondence. (e-mail: jinqiang@upc.edu.cn)
Origin and source characterization of methane
in the shallow-water environment of Southern
Lake Tanganyika Rift Basin, Tanzania
Januarius Matata Bishanga and Jin Qiang*
School of Geoscience, China University of Petroleum, Qingdao 266000, China Lake Tanganyika, located in the western part of Tanzania between 3°S and 9°S lat, harbours
hydro-carbon and non-hydrohydro-carbon gases in its northern and southern shallow-water environment. In this study, a geochemical analysis of stable carbon and hy-drogen isotopes and an interpretation of individual gas molecular composition were done in order to determine the origin and composition of the naturally occurring hydrocarbon gases in Tanganyika Basin. Nitrogen, a non-hydrocarbon gas is a major compo-nent (76.69%–78.31%) by volume percentage followed by methane (11.68%–12.94%) and other higher hydro-carbons (0.16%–1.63%). The isotopic composition of carbon δ 13C
1 and hydrogen δDC1 ranges from –65.32‰ to –65.81‰ and –272.5‰ to –275.9‰ respec-tively. The isotopic compositions of ethane (δDC2 = –36.7‰ to –35.2‰) and propane (δ DC3 = –31.3‰ to –27.5‰) reflect the thermogenic origin of these higher hydrocarbons. According to molecular characterization of carbon and hydrogen isotope ratios and δD-values, methane gas falls in the biogenic origin category and is formed by carbon dioxide reduction. The isotopic composition of CO2 varies between –8.6‰ and –3.4‰. CO2 reduction is also regarded as a mechanism of bio-genic methane formation based on carbon isotope fractionation factors (greater than 0.16).
Keywords: Biogenic origin, isotopic composition,
methane, molecular characterization, shallow-water envi-ronment.
METHANE (CH4) gas venting in Lake Tanganyika,
Tanza-nia has drawn the attention of several researchers, debat-ing on its origin and mechanism of occurrence. The origin, molecular characterization and maturity evalua-tion of hydrocarbon gases in lacustrine settings have long been propounded using hydrocarbon molecular content (methane/(ethane + propane); C1/C2+ or C1/(C2 + C3);
commonly termed the Bernard parameter) and isotope carbon composition (δ 13C
1)1–5. These gases are composed
mainly of methane (C1) with minor amounts of ethane
(C2), higher hydrocarbons (C3+), nitrogen (N2) and carbon
dioxide (CO2). The gases can be produced through
bio-genic (microbially-derived) or thermobio-genic (petroleum-derived) processes6. The significant range in carbon and
hydrogen molecular composition and stable isotope com-position ratios of hydrocarbon gases render interpretative details regarding the gases and their concentration in lacustrine basins depending on environmental settings. Globally, hydrothermally active regions are characte-rized by traces of gas anomalies in the freshwater/ seawater (plumes) or in the sediments7–10. Tropical lakes
accumulate approximately one-quarter of the Earth’s freshwater11 and bear a substantial amount of
hydrocar-bons like methane and non-hydrocarhydrocar-bons like CO2
(greenhouse gases). Most East African rift lakes are hy-drothermally active and contribute an enormous amount of carbon in the lakes’ water column. Till date, the con-tribution of these rift lakes of CH4 gas to the atmosphere
is unknown12. The Ocean Drilling Programme in Middle
Valley, Juan De Fuca Ridge, Leg 1391, East African rift lakes showed the presence of thermogenic gases in the area. The evidence for thermogenic hydrocarbons inclu-ded elevated contents of higher hydrocarbons (δ 13C
1 =
–30.0‰ to –45.0‰) and based on the geologic setting, it is highly probable that the thermogenic gases were formed by hydrothermal processes. However, there is possible evidence of a bacterial gas component admixed with the thermogenic gases. The biogenic hydrocarbons in the free gas are concentrated below the thermocline where bacterial sulphate reduction has removed most or all of the dissolved sulphate3.
Schoell et al.13 discussed the origin of
significantly-high methane gas concentration in Lake Kivu, north of Lake Tanganyika along the rift system, stating that methane gas (C1) was formed by bacterial activity
through acetate fermentation and CO2 reduction. Further
they mentioned that along Lake Kivu and the northern trough of Lake Tanganyika, thermal springs exist and gases vent through the water column, which is associated with both biogenic and hydrothermal origin13–18. Climate
variability, dry and hot conditions lead to decreased lake level by evaporation and changes in aqueous CO2
concen-tration. This phenomenon coupled with increased pH and salinity led to reduced dissolved CO2 concentration19.
Botz and Stoffers7 suggested that active hydrothermalism
occurs in the shallow-water environment of northern Lake Tanganyika. The major gaseous component of magmatic origin is carbon dioxide (δ 13C = –9.1‰ to
–3.8‰). Other variable light hydrocarbons were also detected. The isotopic composition of methane (δ 13C
1 =
–58.0‰ to –50.4‰; δDC1 = –199‰ to –149‰) indicated
both a bacterial and thermogenic source for the hydro-thermal gas. The δDC1 values, however, indicated that
most of the bacterial methane is likely formed by CO2
reduction. Rudd20 also carried out significant work on
methane oxidation in Lake Tanganyika.
Various researchers have developed different hypo-theses to explain the origin of these gases. According to Deines21, Rohrback et al.22 and Chung et al.23, the carbon
isotope (δ 13C) composition of sedimentary organic
matter in geologic age ranges between –35‰ and –20‰. The carbon isotope composition of methane derived from the thermogenic decomposition of organic matter ranges between –50‰ and –20‰, whereas abiogenic methane has δ 13C values between –110‰ to –50‰ (refs 3 and 5). The
carbon isotopic compositions of methane and ethane, are used to determine the kerogen type of gas source rocks24–29.
Microbial hydrocarbons are usually generated from complex hydrocarbons, shallow-depth environments and reasonably low temperatures (<80°–100°C). Also, their stable carbon isotope composition ratios increase with in-creasing carbon number (C1δ 13C < C2δ 13C < C3δ 13C)30,31.
Further distinct characteristics of biogenic gases include low concentrations of higher C2+ hydrocarbons (low C2+/C1)
and isotopically δ 13C light methane (C
1 < –60‰)32,33.
In deep lakes like Tanganyika with anoxic bottom waters, methane oxidizes through a narrow zone in the upper boundary during water stratification. The methane oxidizers consume all the methane as it diffuses through the upper boundary. This phenomenon leads to low con-centration in the oxic water layer and increase in the anoxic zone17,20. Tanganyika rift basin also receives a large
amount of terrestrial organic matter; therefore, the gener-ation of abundant nitrogen gas from the decomposition of organic matter in the Lake Tanganyika is inevitable. Comprehensive geochemical works have been carried out to determine the origin of methane in northern Tan-ganyika Basin7,13,34,35. This study, therefore, focuses on
gas samples collected in two different seasons in the southern Tanganyika Basin. An extensive and concise geochemical interpretation of individual gas molecular composition and isotopes will offer empirical evidence on the origin and composition of the naturally occurring hydrocarbon gases in the southern Tanganyika Basin.
Lake Tanganyika hydrology and geology
Lake Tanganyika, situated in the western region of Tan-zania between 3°S and 9°S latitude at an altitude of about 773 m amsl, is a young (9–12 Ma), half-graben rift lake
within the western branch of the Eastern Africa Rift System. Among the rift system lakes, it is the deepest (~1.47 km) and the largest (34,000 sq. km) in terms of con-tained volume of freshwater in the world, after Lake Baikal in the south of Russia20. The lake extends NW-SE and runs
from north near Lake Kivu to the south near Lake Nyasa. The Lake borders with the Democratic Republic of Congo (DRC) on the west and Zambia on its southern shore. The Lake is meromictic and below the depth of 150 m, the waters and sediments are permanently stratified, anoxic, containing hydrogen sulphide, ammonia and phosphates14,16,36. The Lake water is chemically mildly
alkaline with a pH of 9 (refs 35, 36). Below the thermoc-line, sulphate concentration is significant and favours the formation of a chemically appropriate environment for anaerobic oxidation of methane.
The Lake and the region around it experience a tropical climate (semi-humid). The late Pleistocene–Holocene sediment accumulation rates are estimated to be 0.4– 0.5 mm/year for the deep waters (>600 m) basinal set-tings and could be less on structural highs16. This makes
Lake Tanganyika a potential site for hydrocarbon explo-rations37. Large amounts of organic-rich sediments have
been deposited in this Lake for sufficient organic matter supply, deep water and anoxic condition. The biological precursors for organic matter are mainly diatoms38.
Strati-fied waters and organic-rich sediments in the Lake con-trol the formation of high-potential source rocks15,36,39,40.
Lake Tanganyika consists of several sub-basins with three main depositional basins, namely the northern basin also called the Rusizi sub-basin, the Kigoma sub-basin and the southern sub-basin, separated by the Ubwari and Kalemie horst blocks respectively. The basins consist of half-grabens, with border faults and an accommodation zone separating the north and south basins37,41. The
depo-sitional basins are normally faulted with more than 4 km of sediments deposited in the deepest parts of the fault boundaries37. The resultant tectonic setting defines
discrete structural forms within the lake, such as border fault margins, littoral platforms, midlake structural highs and axial-deep basins42,43.
Studies show that Lake Tanganyika continuously expe-riences an increase in gross and organic carbon (OC) sedimentation due to: (i) enhanced internal upwelling as a result of hydrological changes; (ii) increased terrestrial nutrient inputs caused by increased anthropogenic activi-ties and/or (iii) modified internal food web36,44.
Materials and method
The method employed in this study was developed after McAuliffe45. The headspace equilibration method was
modified to increase sensitivity and storage of samples prior to analysis. The method is designed for open-ocean surface water, freshwater lakes, rivers and groundwater samples collection.
Materials
Major reagents used were potassium hydroxide (KOH) and sodium hydroxide (NaOH) pellets. The 125 ml glass serum bottles used for the headspace equilibration were from the organic geochemistry laboratory of the China University of Petroleum. Black and blue thick butyl rub-ber stoppers (1 cm) that sealed the glass serum bottles were also from the University laboratory. The plastic syringes used were 3 ml, luer lock syringes. Stopcocks were all three-way, polycarbonate, Luer-fitted. Needles were 21G or 22G Precision Glide-type needles. The ultra-high purity grade helium gas, KOH and NaOH pellets were all of the analytical grade and purchased from Bei-jing Chemical Company in China.
Experimental procedure
Eleven gas samples were collected from shallow water (2–33 m water depth) in the southern Tanganyika Basin (Figure 1) and sampled using a rosette equipped with Niskin bottles. After retrieval, the bottles were subsampled immediately to minimize any equilibration of water con-tained in the Niskin bottles with air. Upon filling, the serum bottles were flushed at least two volumes of water to withdraw air bubbles, prevent contamination, and ensure that the water in the bottle was not exposed to air. Next, 0.5 ml of 8 mol l–1 KOH solution was immediately
added to the bottles using a 1 ml syringe inserted as far as possible in the bottle to avoid air bubbles and contamina-tion and which contained no traces of methane. The KOH solution was prepared at least a day prior to insertion into the bottle to let it equilibrate with air. However,
Figure 1. Schematic map of Lake Tanganyika showing the sampling
sites T1, T2 and T3 in the southern depositional basin. Kigoma and Rusizi (Ruzizi) are the other depositional basins.
it is advised to use KOH solution as opposed to adding KOH pellets directly to the sample to avoid contamina-tion with methane. The bottles were then capped without headspace using butyl rubber stoppers probed with a needle connected to a 3 ml syringe to allow air and water to leave the bottle when the stopper was inserted. The re-sidual water in the syringe blocked back-injection of air when the needle was withdrawn. This also allowed water to expand or contract as the temperature of the sample equilibrates to laboratory temperature.
Once all Niskin bottles were subsampled and the sam-ples had attained room temperature, the needles and syringes were withdrawn and the stoppers secured with aluminium crimps. To prevent bottle cracking due to temperature variations, a 10 ml headspace of methane-free helium was added to the bottles using a 10 ml sy-ringe, a 21G or 22G needle and a three-way stopcock. At the same time, 10 ml water was removed with a second syringe, needle and stopcock assembly. Once the head-space was added, the bottles were stored upside down to isolate gas from the stopper until analysis of the head-space methane. All the geochemical and stable carbon isotope composition analyses were performed using stan-dard techniques as described by Schoell46,47. The methane
carbon isotopes were determined using the method of Malanotte-Rizzoli et al.48, and results denoted in the
common δ-notation relative to Vienna Pee Dee Belemnite (VPDB). The methane concentrations were determined on a gas chromatograph (Agilent, 6890). Standard deviation (SD) and standard error (SE) were calculated to deter-mine the consistency of the results. The SD of sample measurements taken in December 2013 and March 2014 was 0.49 and 0.45, while SE was 0.22 and 0.19 respec-tively. When all datasets were combined, the SD and SE values were 0.47 and 0.14 respectively.
Results
Table 1 provides the location and approximate sampling depth for gas samples in the southern Tanganyika basin. Table 2 shows the results of the gas chromatographic analysis, which is the percentage of the various extracted gases (hydrocarbons and non-hydrocarbons) by volume assuming that the total is 100%. The stable isotope data indicate methane as the predominant hydrocarbon gas present. The percentage by volume of C1 varies
between 11.68 and 12.94, while the minimum amount of
Table 1. Locations (coordinates) and approximate water depths of
gas samples in southern Tanganyika Basin Station
Approximate
depth (m) Easting (UTM) Northing (UTM)
T1 8 193,296 9,231,144
T2 15 238,020 9,147,091
Table 2. Extracted gases percentage by volume in southern Lake Tanganyika Percentage Sample ID Ar O2 CO2 N2 C1 C2 C3 C1/(C2 + C3) T1-D-500-1 1.5 6.3 0.52 78.31 12.12 0.94 0.31 9.70 T1-D-500-2 1.6 6.4 0.54 77.53 12.64 1.05 0.24 9.80 T2-D-500-3 1.5 7.3 0.43 77.82 11.71 0.73 0.51 9.44 T3-D-500-2 1.6 7.1 0.52 77.85 11.68 0.84 0.41 9.34 T1-D-400-1 1.6 5.3 0.53 78.2 12.71 1.21 0.45 7.66 T1-D-400-1 1.52 6.55 0.55 77.12 12.94 0.98 0.34 9.80 T1-D-400-3 1.58 7.21 0.51 77.26 12.29 0.86 0.29 10.69 T2-C-200-1 1.53 6.95 0.49 76.69 12.9 0.79 0.65 8.96 T2-D-400-2 1.6 6.42 0.5 78.22 11.92 0.72 0.62 8.90 T3-D-500-1 1.48 7.1 0.55 77.84 11.95 0.92 0.16 11.06 T3-D-500-3 1.51 6.57 0.58 76.96 12.64 1.63 0.11 7.26 Ar, Argon; O2, Oxygen; CO2, Carbon dioxide; N2, Nitrogen; C1, Methane; C2,
Ethane: C3, Propane.
Table 3. Isotopic composition of methane (δ 13C
1, δDC1), ethane (δ 13C–C2), propane (δ 13C–C3) and carbon dioxide
(δ 13C–CO
2) from southern Lake Tanganyika
δ 13C
1 δDC1 δ 13C–C2 δ 13C–C3 δ 13C–CO2
Sample ID (‰ versus PDB) (‰ versus VSMOW) (‰ versus PDB) (‰ versus PDB) (‰ versus PDB)
T1-D-500-1 –65.40 –274.4 –35.4 –30.1 –8.3 T1-D-500-2 –65.46 –272.5 –36.6 –30.3 –6.8 T2-D-500-3 –65.75 –274.7 –35.2 –29.8 –5.7 T3-D-500-2 –65.73 –275.7 –35.7 –29.3 –3.4 T1-D-400-1 –65.77 –274.8 –35.3 –29.3 –4.7 T1-D-400-1 –65.32 –274.5 –35.2 –30.6 –8.6 T1-D-400-3 –65.57 –272.7 –36.1 –31.3 –6.1 T2-C-200-1 –65.45 –274.6 –35.8 –29.4 –4.7 T2-D-400-2 –65.48 –275.9 –36.7 –28.3 –3.8 T3-D-500-1 –65.60 –274.9 –35.9 –29.8 –4.1 T3-D-500-3 –65.81 –274.6 –36.6 –27.5 –3.7
VSMOW, Vienna Standard Mean Ocean Water – an isotopic water benchmark used for both hydrogen and oxygen isotopes; PDB, Pee Dee Belemnite – for carbon isotopes in organic matter.
higher hydrocarbons ethane to propane varies from 0.16 to 1.63. The gas ratios C1/(C2 + C3), i.e. methane/
(ethane + propane) range from 7.26 to 11.06 (Table 2). The C1/(C2 + C3) ratio is an indicator of the origin of the
gases and decreases with depth towards the Basin depo-centres. Most researchers relate these trends to the deeper generation of thermogenic wet gases while at shallower depths, late-stage microbial gas is subject to meteoric recharge1,49,50. The isotopic composition of carbon δ 13C
1
and hydrogen δDC1 ranges from –65.32‰ to –65.81‰
and –272.5‰ to –275.9‰ respectively (Table 3). The isotopic composition of carbon dioxide varies between –8.6‰ and –3.4‰ (Table 4), falling in the range of magmatic origin. However, thermogenic CO2 should be
less in carbon isotope as it is derived from decarboxyla-tion reacdecarboxyla-tions of organic matter2,13.
Discussion
Methane, carbon and hydrogen isotope ratios
Natural gas, also called fossil gas, is found in a wide variety of environments. Methane is a major constituent
of natural gas, but the latter also contains heavier hydro-carbons (ethane, propane, butane), small amounts of nitrogen, carbon dioxide, hydrogen sulphide, rare gases and trace amounts of water. Natural gas can be generated through two processes, viz. microbial and thermogenic. The carbon δ 13C
1 and hydrogen δ DC1 isotopes and
mo-lecular compositions are used to determine the methane gas origin and generation pathways. The most important parameters used to distinguish the gas types are their iso-topic composition of stable carbon δ 13C
1 and hydrogen
δ DC1 along with the relative abundance of methane to
other hydrocarbon gases2,3,5,33,47,51. The different types of
methane have characteristic carbon and hydrogen isotope compositions, which vary with source type and maturity in the case of thermogenic methane5. The carbon isotope
compositions of methane derived from the primary ther-mogenic decomposition of organic matter ranges between –50‰ and –20‰, whereas biogenic methane has δ 13C
1
values between –110 and –50‰ and less than –60‰ for carbon dioxide reduction pathway3,5,49,52. An intermediate
methane carbon isotope composition between –50‰ and –60‰ may be interpreted as mixing origins between microbial and thermogenic gases49,53. The combination of
carbon and hydrogen isotope analysis of natural gas is therefore an important approach to decipher the different origins of gases.
The hydrocarbon composition of methane can also be quantified in a number of ways, such that it is important to precisely distinguish between gas dryness and wetness. The ratio of methane to the sum of ethane and propane (C1/(C2 + C3)) is extensively used to distinguish between
microbial and thermogenic gases. The ratio of methane to the sum of ethane and propane greater than 1000 and less than 100 (wet gases analogous with oil show (C1/(C2 + C3)) < 50) is an indicator of microbial and
thermogenic gas respectively, when used in combination with methane carbon isotope composition3,49–51,54,55.
Table 3 shows the isotopic composition of methane (δ 13C
1, δ DC1), ethane (δ 13C–C2), propane (δ 13C–C3),
and carbon dioxide (δ 13C–CO
2) from the southern
Tan-ganyika Basin. The isotopic composition of carbon δ 13C 1
and hydrogen δ DC1 ranges from –65.32‰ to –65.81‰
and –272.5‰ to –275.9‰ respectively. Various diagrams previously developed for the classification of natural gas-es, including the Schoell5 and Whiticar51 diagrams, have
been used to determine the origin of methane. The most widely used is the Bernard diagram (modified after Ber-nard et al.54), which compares the molecular ratio of
C1/(C2 + C3) with δ 13C–CH4 to distinguish between
mi-crobial (biogenic) and thermogenic gases. Figure 2 displays the δ 13C
1 against δ DC1 plot
com-prising isotopic data for methane from southern Lake Tanganyika. According to the original diagram (refer to Whiticar56), methane gas from Tanganyika Basin falls in
the category of bacterial/biogenic origin.
Figure 3 shows the isotopic composition of biogenic methane within the isotopic fields of CO2 reduction.
Therefore, CO2 reduction is a controlling mechanism
of biogenic methane formation in Lake Tanganyika. However, according to the Bernard diagram (Figure 4), methane in southern Lake Tanganyika falls in the region of mixed origin (biogenic and thermogenic) which agrees with the relative amounts of higher hydrocarbons (C2 and
C3) by comparing the molecular (C1/(C2 + C3)) ratio with
Table 4. Isotopic composition of carbon dioxide from
shallow waters in southern Lake Tanganyika Sample ID CO2 (%) δ 13C–CO2 (‰ versus PDB)
T1-D-500-1 0.52 –8.3 T1-D-500-2 0.54 –6.8 T2-D-500-3 0.43 –5.7 T3-D-500-2 0.52 –3.4 T1-D-400-1 0.53 –4.7 T1-D-400-1 0.55 –8.6 T1-D-400-3 0.51 –6.1 T2-C-200-1 0.49 –4.7 T2-D-400-2 0.50 –3.8 T3-D-500-1 0.55 –4.1 T3-D-500-3 0.58 –3.7 the δ 13C–CH
4 isotopic composition. Despite the presence
of minimum amount of higher hydrocarbons, i.e. ethane, propane, etc. the source interpretation is confirmed by the Bernard diagram54. The isotopic composition of ethane
(δ DC2 = –36.7‰ to –35.2‰) and propane (δ DC3 =
–31.3‰ to –27.5‰) reflect the thermogenic origin of these higher hydrocarbons. This makes it hard to rule out methane origin from thermogenic processes as well. The generation of N2 and methane may have occurred by the
process of thermal decomposition of humic organic matter, based on the quantity of N2 gas, while the most
important methane-producing pathways in lacustrine sediments are biogenic (methanogenesis from acetate) and CO2 reduction57.
The fermentation process in shallow deposited sedi-ments contributes to the formation of biogenic methane in Lake Tanganyika and both processes (fermentation and CO2 reduction) may be involved. Bacterial oxidation
Figure 2. Plot of δ 13C
1 versus δ DC1 showing the isotopic
composi-tion of methane in southern Lake Tanganyika.
Figure 3. δ 13C
1 against δ DC1 plot showing the isotopic composition
of methane from southern Lake Tanganyika (modified from Schoell47
significantly influences the production of methane rather than lighter hydrocarbons.
Carbon
δ
13C1 isotopic characterization
Based on primary processes, the δ 13C
1 isotopic
characte-rization of natural gases is related to the three main stages of thermal evolution of organic matter in sedimentary rocks, i.e. metagenetic (–40‰ to –25‰), catagenetic (–55‰ to –40‰) and diagenetic (–90‰ to –55‰).
Carbon dioxide carbon isotope composition
The possible sources of CO2 in lake waters and sediments
may include microbial degradation of organic substrates, dissolved CO2 from the atmosphere, magmatic/mantle
origin, thermal decomposition of carbonates, and thermal maturation of kerogen49,58,59. The best way to evaluate the
origin of CO2 is to assess the relationship between
con-centration and carbon isotopic composition. Gases with CO2 content greater than 10 vol%, mostly have isotopic
composition with δ 13C
1 values between –3‰ and –10‰,
which fall in the range of CO2 from geothermal
(magmat-ic) systems and are predominantly of magmatic/mantle origin. Gases with CO2 content less than 10 vol% display
a wide range of δ 13C values. Gases with CO
2 content less
than 10 vol% and isotopic composition in the inorganic range between –3‰ and –10‰ are probably thermogenic or mixtures and are affected by microbial degrada-tion49,53. In this study, the isotopic composition of carbon
dioxide varies between –8.6‰ and –3.4‰ (Table 4), falling in the range of CO2 from geothermal (magmatic)
systems (Figure 5).
During methanogenesis, the covariance of δ 13C
1 values
of methane and carbon dioxide is presumed where
Figure 4. Schematic illustration of Bernard diagram for southern
Lake Tanganyika gas samples modified from Schoell47 and Bernard
et al.54.
methanogens utilize the CO2 reduction pathway. The CO2
and CH4 isotopic composition values associated with
carbonate reduction range from –49‰ to over –100‰. However, the most frequently observed values fall between –60‰ and –80‰. For acetate fermentation and methano-genesis utilizing methylated substrates, the CO2 and CH4
isotopic composition values are relatively low, with aver-age values between –40‰ and –55‰ (refs 2, 49, 51 and 60). A plot of δ 13C-CO
2 against δ 13C-CH4 values shows
the carbon isotope fractionation lines for various methano-genic pathways, where αCO2–CH4 = (1000 + δ 13C–CO2)/
(1000 + δ 13C–CH
4)49. The carbon isotope fractionation
values between 1.03 and 1.06, 1.06 and 1.09 are characteristic of acetate and methyl-type fermentation
Figure 5. Carbon isotopic composition of CO2 versus molecular
per-centage of southern Lake Tanganyika methane. Organic – from thermal decomposition of organic matter; biogenic – influenced by microbial activity and inorganic – from igneous input (magmatic systems).
Figure 6. Plot of carbon dioxide δ 13C versus methane δ 13C values
showing the carbon isotope fractionation lines for various methanogenic pathways in southern Lake Tanganyika.
and CO2 reduction respectively49–51. This demonstrates
that the carbon isotope fractionation between CO2 and
CH4 (Figure 6) may as well be used to depict and
distin-guish between gases of microbial and thermogenic origin (Figure 6). The isotope fractionation factors reveal that CO2 reduction has been implicated in gas generation in
Lake Tanganyika.
In this study, N2 percentage volume is higher in all
samples (Table 2) and presumed to be the major gas component (76.69%–78.31%) in the Tanganyika Basin. This significant quantity suggests fermentation of autoch-thonous organic matter and deposition of clastic sediments in shallow depths. Water-column stratification and upwel-ling are important processes that introduce ammonia (NH3)
to the upper southern end water surfaces and assumed to be assimilated by planktons. A large proportion of NH3
introduced into the layer is lost either to the atmosphere or by oxidation to N2 in solution39. It can be concluded that
thermocline and upwelling are significant processes for rapid oxidation to nitrogen in the southern Basin.
Conclusion
In addition to light hydrocarbon methane, southern Lake Tanganyika Basin contains sufficient quantities of other gases (N2, O2 and Ar). The isotopic composition of
methane in Lake Tanganyika indicates biogenic origin with carbon δ 13C
1 and hydrogen δDC1 isotopic
composi-tion ranging from –65.32‰ to –65.81‰ and –272.5‰ to –275.9‰ respectively, while the gas ratios C1/(C2 + C3)
range from 7.26 to 11.06 of mixed origin. The isotopic compositions of ethane (δ DC2 = –36.7‰ to –35.2‰) and
propane (δDC3 = –31.3‰ to –27.5‰) reflect the
thermo-genic origin of these higher hydrocarbons. CO2 reduction
and fermentation are the processes for biogenic methane formation. The possible source of CO2 in lake waters is
magmatic/mantle origin and its isotopic composition varies between –8.6‰ and –3.4‰. Moreover, the isotope fractionation factor values which fall above 1.06 show that CO2 reduction has been implicated in gas generation
in the Tanganyika. There is evidence that more CH4 is
produced from enormous autochthonous (derived from aquatic plants and phytoplankton) organic matter present in the deep southern Tanganyika Basin than from alloch-thonous (derived from land plants) organic matter. A high organic matter supply rate and accumulation can stimu-late significant sediment CH4 formation and emission,
especially for biologically reactive organic matter. Ther-mal decomposition of organic matter and oxidation of NH3 to N2 may have been the reason for N2, CH4 and CO2
gases in the southern Tanganyika Basin.
1. Rice, D. D. and Claypool, G. E., Generation, accumulation, and resource potential of biogenic gas. Am. Assoc. Pet. Geol. Bull., 1981, 65, 5–25.
2. Whiticar, M. J., Faber, E. and Schoell, M., Biogenic methane for-mation in marine and freshwater environments: CO2 reduction
versus acetate fermentation – isotope evidence. Geochim. Cosmo-chim. Acta, 1986, 50, 693–709.
3. Whiticar, M. J., Faber, E., Whelan, J. K. and Simoneit, B. R., Thermogenic and bacterial hydrocarbon gases (free and sorbed) in Middle Valley, Juan de Fuca Ridge, Leg 139. In Proceedings of the Ocean Drilling Program, Scientific Results, Texas A and M University, TX, United States, 1994, pp. 467–477.
4. Yen, T. F. and Chilingar, G. V., A Review of: B. P. Tissot and D. H. Welte, Petroleum Formation and Occurrence. A New Approach to Oil and Gas Exploration, Springer-Verlag, Berlin, 1978, 538 pp. 243 Figures. Energy Sources, 1979, 4, 367–382.
5. Schoell, M., The hydrogen and carbon isotopic composition of methane from natural gases of various origins. Geochim. Cosmo-chim. Acta, 1980, 44, 649–661.
6. Tissot, B. and Welte, D., Petroleum Formation and Occurrence, Springer-Verlag, Berlin, Germany, 1984.
7. Botz, R. and Stoffers, P., Light hydrocarbon gases in Lake Tanga-nyika hydrothermal fluids (East-Central Africa). Chem. Geol., 1993, 104, 217–224.
8. Welhan, J. T. and Craig, H. J., Methane and hydrogen in East Pacific Rise hydrothermal fluids. Geophys. Res. Lett., 1979, 6, 829–831.
9. Whiticar, M. J. and Suess, E. J., Hydrothermal hydrocarbon gases in the sediments of the King George Basin, Bransfield Strait, Antarctica. Appl. Geochem., 1990, 5, 135–147.
10. Welhan, J. and Lupton, J., Light hydrocarbon gases in Guaymas Basin hydrothermal fluids: thermogenic versus abiogenic origin. Am. Assoc. Pet. Geol. Bull., 1987, 71, 215–223.
11. Bootsma, H. A. and Hecky, R. E. J., A comparative introduction to the biology and limnology of the African Great Lakes. J. Great Lakes Res., 2003, 29, 3–18.
12. Durisch‐Kaiser, E. et al., What prevents outgassing of methane to the atmosphere in Lake Tanganyika? J. Geophys. Res., 2011, 116. 13. Schoell, M., Tietze, K. and Schoberth, S., Origin of methane in
Lake Kivu (east-central Africa). Chem. Geol., 1988, 71, 257–265. 14. Coulter, G. W., Hydrological changes in relation to biological
production in Southern Lake Tanganyika. Limnol. Oceanogr., 1963, 8, 463–477.
15. Craig, H. (ed.), Lake Tanganyika geochemical and hydrographic study: 1973 expedition. In Scripps Institution of Oceanography, University of California, San Diego, California, USA, p. 83. 16. Degens, E. T., Von Herzen, R. P. and Wong, H.-K., Lake
Tanga-nyika: water chemistry, sediments, geological structure. Naturwis-senschaften, 1971, 58, 229–241.
17. Degens, E. T., von Herzen, R. P., Wong, H.-K., Deuser, W. G. and Jannasch, H. W., Lake Kivu: structure, chemistry and biology of an East African rift lake. Geol. Rund., 1973, 62, 245–277. 18. Tiercelin, J.-J. et al., Hydrothermal vents in Lake Tanganyika,
East African, Rift system. Geology, 1993, 21, 499–502.
19. Leng, M. J. et al. (eds), Isotopes in lake sediments. In Isotopes in Palaeoenvironmental Research, Springer, 2006, pp. 147–184. 20. Rudd, J. W., Methane oxidation in Lake Tanganyika (East Africa).
Limnol. Oceanogr., 1980, 25, 958–963.
21. Deines, P., The isotopic composition of reduced organic carbon. In The Terrestrial Environment (eds Fritz, P. and Fontes, J. C.), Elsevier, Amsterdam, The Netherlands, pp. 329–406.
22. Peters, K., Rohrback, B. and Kaplan, I., Geochemistry of artifi-cially heated humic and sapropelic sediments – I: Protokerogen. AAPG Bull., 1981, 65, 688–705.
23. Chung, H. M., Sackett, W. M. J. P. and Earth, C. O. T., Carbon isotope effects during the pyrolytic formation of early methane from carbonaceous materials. Phys. Chem. Earth, 1980, 12, 705–710. 24. Ni, Y. et al., Stable hydrogen and carbon isotopic ratios of coal-derived and oil-coal-derived gases: a case study in the Tarim basin, NW China. Int. J. Coal Geol., 2013, 116, 302–313.
25. Ping, H., Chen, H., Zhu, J., George, S. C., Mi, L., Pang, X. and Zhai, P., Origin, source, mixing, and thermal maturity of natural
gases in the Panyu lower uplift and the Baiyun depression, Pearl River Mouth Basin, northern South China Sea. Am. Assoc. Pet. Geol. Bull., 2018, 102, 2171–2200.
26. Cramer, B., Faber, E., Gerling, P. and Krooss, B. M., Reaction kinetics of stable carbon isotopes in natural gas: insights from dry, open system pyrolysis experiments. J. Energy Fuels, 2001, 15, 517–532.
27. Hao, F., Guo, T., Zhu, Y., Cai, X., Zou, H. and Li, P., Evidence for multiple stages of oil cracking and thermochemical sulfate re-duction in the Puguang gas field, Sichuan Basin, China. Am. As-soc. Pet. Geol. Bull., 2008, 92, 611–637.
28. Hoşgörmez, H., Yalçın, M. N., Cramer, B., Gerling, P. and Mann, U., Molecular and isotopic composition of gas occurrences in the Thrace basin (Turkey): origin of the gases and characteristics of possible source rocks. J. Chem. Geol., 2005, 214, 179–191. 29. Laughrey, C. D. and Baldassare, F. J., Geochemistry and origin of
some natural gases in the Plateau Province, central Appalachian Basin, Pennsylvania and Ohio. Am. Assoc. Pet. Geol. Bull., 1998,
82, 317–335.
30. Chung, H., Gormly, J. and Squires, R. J., Origin of gaseous hydro-carbons in subsurface environments: theoretical considerations of carbon isotope distribution. J. Chem. Geol., 1988, 71, 97–104. 31. McCarty, H. B. and Felbeck Jr, G. T., High temperature
simula-tion of petroleum formasimula-tion – IV. Stable carbon isotope studies of gaseous hydrocarbons. J. Org. Geochem., 1986, 9, 183–192. 32. Machel, H., Bacterial and thermochemical sulfate reduction in
diage-netic settings – old and new insights. J. Sediment. Geol., 2001, 140, 143–175.
33. Schoell, M., Genetic characterization of natural gases. Am. Assoc. Pet. Geol. Bull., 1983, 67, 2225–2238.
34. Deuser, W., Degens, E., Harvey, G. and Rubin, M., Methane in Lake Kivu: new data bearing on its origin. Science, 1973, 181, 51–54.
35. Gourgue, O., Deleersnijder, E., White, L., Toward a generic method for studying water renewal, with application to the epilim-nion of Lake Tanganyika. J. Estuar. Coastal Shelf Sci., 2007, 74, 628–640.
36. Cohen, A. S. et al., Paleolimnological investigations of anthropo-genic environmental change in Lake Tanganyika: IX. Summary of paleorecords of environmental change and catchment deforestation at Lake Tanganyika and impacts on the Lake Tanganyika ecosys-tem. J. Paleolimnol., 2005, 34, 125–145.
37. Rosendahl, B. et al., Structural expressions of rifting: lessons from Lake Tanganyika, Africa. Geol. Soc., London, Spec. Publ., 1986,
25, 29–43.
38. Huc, A. et al. (eds), Northern Lake Tanganyika: an example of organic sedimentation in an anoxic rift lake. In Laustrine Basin Exploration: Case Studies and Modern Analogs, AAPG Memoir, 1990, pp. 169–208.
39. Edmond, J., Stallard, R., Craig, H., Craig, V., Weiss, R. and Coulter, G., Nutrient chemistry of the water column of Lake Tanganyika. Limnol. Oceanogr., 1993, 38, 725–738.
40. Demaison, G. J. and Moore, G. T., Anoxic environments and oil source bed genesis. AAPG Bull., 1980, 64, 1179–1209.
41. Quennel, A., McKinley, A. and Aitken, W., Summary of the Geology of Tanganyika, Memoir I, Part II: Introduction and Stratigraphy, Government Printer, Dar es Salaam, Tanzamia, 1956, p. 264.
42. Tiercelin, J., Cohen, A., Soreghan, M. and Lezzar, K., Pleisto-cene–modern deposits of the Lake Tanganyika Rift Basin, East Africa: a modern analog for lacustrine source rocks and reservoirs. In SEPM Core Workshop, 1994, vol. 19, pp. 37–59.
43. Tiercelin, J.-J., Soreghan, M., Cohen, A. S., Lezzar, K.-E. and Bouroullec, J.-L., Sedimentation in large rift lakes: example from the Middle Pleistocene – modern deposits of the Tanganyika
Trough, East African Rift system. Bull. Centres Rech. Explor.-Prod. Elf Aquitaine, 1992, 16, 83–111.
44. Pasche, N. et al., Abrupt onset of carbonate deposition in Lake Kivu during the 1960s: response to recent environmental changes. J. Paleolimnol., 2010, 44, 931–946.
45. McAuliffe, C., GC determination of solutes by multiple phase equilibration. A new general technique. Chem. Tech., 1971, 1, 46–51. 46. Schoell, M., Recent advances in petroleum isotope geochemistry.
Org. Geochem., 1984, 6, 645–663.
47. Schoell, M., Multiple origins of methane in the Earth. Chem. Geol., 1988, 71, 1–10.
48. Malanotte-Rizzoli, P. et al., A synthesis of the Ionian Sea hydro-graphy, circulation and water mass pathways during POEM-Phase I. Prog. Oceanogr., 1997, 39, 153–204.
49. Golding, S. D., Boreham, C. J. and Esterle, J. S., Stable isotope geochemistry of coal bed and shale gas and related production waters: a review. Int. J. Coal Geol., 2013, 120, 24–40.
50. Flores, R. M., Rice, C. A., Stricker, G. D., Warden, A. and Ellis, M. S., Methanogenic pathways of coal-bed gas in the Powder River Basin, United States: the geologic factor. Int. J. Coal Geol., 2008,
76, 52–75.
51. Whiticar, M. J., Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane. Chem. Geol., 1999,
161, 291–314.
52. Strąpoć, D. et al., Biogeochemistry of microbial coal-bed methane. Annu. Rev. Earth Planet. Sci., 2011, 39, 617–656. 53. Boreham, C. J., Edwards, D. S., Hope, J. M., Chen, J. and Hong,
Z., Carbon and hydrogen isotopes of neo-pentane for biodegraded natural gas correlation. Org. Geochem., 2008, 39, 1483–1486. 54. Bernard, B. B., Brooks, J. M. and Sackett, W. M., Light
hydrocar-bons in recent Texas continental shelf and slope sediments. J. Geophys. Res.: Oceans, 1978, 83, 4053–4061.
55. Osborn, S. G. and McIntosh, J. C., Chemical and isotopic tracers of the contribution of microbial gas in Devonian organic-rich shales and reservoir sandstones, northern Appalachian Basin. Appl. Geochem., 2010, 25, 456–471.
56. Whiticar, M. J., A geochemial perspective of natural gas and atmospheric methane. Org. Geochem., 1990, 16, 531–547.
57. Conrad, R., Quantification of methanogenic pathways using stable carbon isotopic signatures: a review and a proposal. J. Organic Geochem., 2005, 36, 739–752.
58. Dai, J. et al., Geochemical characteristics of the abiogenic alkane gases in the Songliao Basin, China. In AAPG Hedberg Research Conference – Natural Gas Geochemistry: Recent Developments, Applications and Technologies, Beijing, China, 2011.
59. Pallasser, R., Recognising biodegradation in gas/oil accumulations through the δ 13C compositions of gas components. Org.
Geo-chem., 2000, 31, 1363–1373.
60. Jenden, P. and Kaplan, I., Comparison of microbial gases from the Middle America Trench and Scripps Submarine Canyon: implica-tions for the origin of natural gas. Appl. Geochem., 1986, 1, 631– 646.
ACKNOWLEDGEMENTS. This work was supported by the Certifi-cate of National Science and Technology Major Project of the Ministry of Science and Technology of China (2016ZX05006-007), National Natural Science Foundation of China (41702139), Natural Science Foundation of Shandong Province (ZR2017BD036) and the Fundamen-tal Research Funds for the Central Universities (18CX02008A). We thank the Tanzania Petroleum Development Co-operation (TPDC) for providing geological datasets.
Received 24 June 2020; revised accepted 20 November 2020 doi: 10.18520/cs/v120/i6/1066-1073
